| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 25.0, | |
| "global_step": 2350, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 3.4338, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.4137182235717773, | |
| "eval_runtime": 36.2713, | |
| "eval_samples_per_second": 5.514, | |
| "eval_steps_per_second": 0.689, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.9565, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.173758029937744, | |
| "eval_runtime": 35.902, | |
| "eval_samples_per_second": 5.571, | |
| "eval_steps_per_second": 0.696, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0, | |
| "loss": 2.7101, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.012174606323242, | |
| "eval_runtime": 35.6264, | |
| "eval_samples_per_second": 5.614, | |
| "eval_steps_per_second": 0.702, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 2.7515, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.964645504951477, | |
| "eval_runtime": 35.4193, | |
| "eval_samples_per_second": 5.647, | |
| "eval_steps_per_second": 0.706, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 2.724, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.1284220218658447, | |
| "eval_runtime": 34.8555, | |
| "eval_samples_per_second": 5.738, | |
| "eval_steps_per_second": 0.717, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 2.6193, | |
| "step": 564 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.9379758834838867, | |
| "eval_runtime": 36.3088, | |
| "eval_samples_per_second": 5.508, | |
| "eval_steps_per_second": 0.689, | |
| "step": 564 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 0.0, | |
| "loss": 2.5032, | |
| "step": 658 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.9285995960235596, | |
| "eval_runtime": 35.9858, | |
| "eval_samples_per_second": 5.558, | |
| "eval_steps_per_second": 0.695, | |
| "step": 658 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 2.5342, | |
| "step": 752 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.9365949630737305, | |
| "eval_runtime": 35.1567, | |
| "eval_samples_per_second": 5.689, | |
| "eval_steps_per_second": 0.711, | |
| "step": 752 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4e-05, | |
| "loss": 2.5519, | |
| "step": 846 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.9736474752426147, | |
| "eval_runtime": 35.9727, | |
| "eval_samples_per_second": 5.56, | |
| "eval_steps_per_second": 0.695, | |
| "step": 846 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.4988, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.881581425666809, | |
| "eval_runtime": 35.6769, | |
| "eval_samples_per_second": 5.606, | |
| "eval_steps_per_second": 0.701, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 2.5101, | |
| "step": 1034 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 1.8453679084777832, | |
| "eval_runtime": 33.4013, | |
| "eval_samples_per_second": 5.988, | |
| "eval_steps_per_second": 0.748, | |
| "step": 1034 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 2e-05, | |
| "loss": 2.4441, | |
| "step": 1128 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.8143038749694824, | |
| "eval_runtime": 34.7625, | |
| "eval_samples_per_second": 5.753, | |
| "eval_steps_per_second": 0.719, | |
| "step": 1128 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 2.3857, | |
| "step": 1222 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 1.7919152975082397, | |
| "eval_runtime": 33.975, | |
| "eval_samples_per_second": 5.887, | |
| "eval_steps_per_second": 0.736, | |
| "step": 1222 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 2.2877, | |
| "step": 1316 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 1.7400457859039307, | |
| "eval_runtime": 34.7007, | |
| "eval_samples_per_second": 5.764, | |
| "eval_steps_per_second": 0.72, | |
| "step": 1316 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 0.0, | |
| "loss": 2.3013, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 1.7408628463745117, | |
| "eval_runtime": 34.5398, | |
| "eval_samples_per_second": 5.79, | |
| "eval_steps_per_second": 0.724, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 2e-05, | |
| "loss": 2.3134, | |
| "step": 1504 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.7698218822479248, | |
| "eval_runtime": 34.7625, | |
| "eval_samples_per_second": 5.753, | |
| "eval_steps_per_second": 0.719, | |
| "step": 1504 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 1.5e-05, | |
| "loss": 2.3423, | |
| "step": 1598 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 1.7581219673156738, | |
| "eval_runtime": 35.8583, | |
| "eval_samples_per_second": 5.578, | |
| "eval_steps_per_second": 0.697, | |
| "step": 1598 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1e-05, | |
| "loss": 2.3536, | |
| "step": 1692 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 1.7658276557922363, | |
| "eval_runtime": 35.6301, | |
| "eval_samples_per_second": 5.613, | |
| "eval_steps_per_second": 0.702, | |
| "step": 1692 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 5e-06, | |
| "loss": 2.2957, | |
| "step": 1786 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 1.7328851222991943, | |
| "eval_runtime": 36.2644, | |
| "eval_samples_per_second": 5.515, | |
| "eval_steps_per_second": 0.689, | |
| "step": 1786 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0, | |
| "loss": 2.274, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 1.7334604263305664, | |
| "eval_runtime": 34.2732, | |
| "eval_samples_per_second": 5.835, | |
| "eval_steps_per_second": 0.729, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 2.2906, | |
| "step": 1974 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 1.734320878982544, | |
| "eval_runtime": 34.5789, | |
| "eval_samples_per_second": 5.784, | |
| "eval_steps_per_second": 0.723, | |
| "step": 1974 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 1.2e-05, | |
| "loss": 2.2492, | |
| "step": 2068 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 1.7080261707305908, | |
| "eval_runtime": 35.1218, | |
| "eval_samples_per_second": 5.694, | |
| "eval_steps_per_second": 0.712, | |
| "step": 2068 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 2.2516, | |
| "step": 2162 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 1.718016505241394, | |
| "eval_runtime": 34.8144, | |
| "eval_samples_per_second": 5.745, | |
| "eval_steps_per_second": 0.718, | |
| "step": 2162 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.2574, | |
| "step": 2256 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 1.7081444263458252, | |
| "eval_runtime": 34.6174, | |
| "eval_samples_per_second": 5.777, | |
| "eval_steps_per_second": 0.722, | |
| "step": 2256 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 0.0, | |
| "loss": 2.2508, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 1.7065184116363525, | |
| "eval_runtime": 34.9055, | |
| "eval_samples_per_second": 5.73, | |
| "eval_steps_per_second": 0.716, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "step": 2350, | |
| "total_flos": 8.960483088e+18, | |
| "train_loss": 0.45198433734001, | |
| "train_runtime": 1195.2772, | |
| "train_samples_per_second": 15.687, | |
| "train_steps_per_second": 1.966 | |
| } | |
| ], | |
| "max_steps": 2350, | |
| "num_train_epochs": 25, | |
| "total_flos": 8.960483088e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |